WO2023058157A1 - Treatment assistance device, system, and method - Google Patents

Treatment assistance device, system, and method Download PDF

Info

Publication number
WO2023058157A1
WO2023058157A1 PCT/JP2021/036999 JP2021036999W WO2023058157A1 WO 2023058157 A1 WO2023058157 A1 WO 2023058157A1 JP 2021036999 W JP2021036999 W JP 2021036999W WO 2023058157 A1 WO2023058157 A1 WO 2023058157A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
store
disease
product
user
Prior art date
Application number
PCT/JP2021/036999
Other languages
French (fr)
Japanese (ja)
Inventor
和寛 檜山
Original Assignee
メディテック株式会社
ケイ－トムト コー リミテッド
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by メディテック株式会社, ケイ－トムト コー リミテッド filed Critical メディテック株式会社
Priority to PCT/JP2021/036999 priority Critical patent/WO2023058157A1/en
Publication of WO2023058157A1 publication Critical patent/WO2023058157A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Definitions

  • the present invention relates to a treatment support device, system, and method for skin diseases, and more particularly to a treatment support device, method, and system using AI image diagnosis that contributes to self-medication.
  • Diagnosis of skin diseases has aspects of image analysis. Inventions using artificial intelligence as systems for analyzing medical images are known, but they are not inventions intended for diagnosing skin diseases (see Patent Document 1).
  • an object of the present invention is to provide a treatment support device, system, and method using AI image diagnosis that contributes to self-medication.
  • a skin disease treatment support device connected to a user terminal via a network, wherein the treatment support device User information including disease images and location information is received from the user terminal, and diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and the product are transmitted to the user terminal.
  • a device comprising [2]
  • the treatment support device is further connected to a store management device via a network, the transmitting/receiving means further transmits store information including inventory information, and the control means further controls the inventory of the product from the store.
  • [4] The device according to any one of [1] to [3], wherein the user terminal is one selected from the group consisting of a smartphone, a tablet terminal, and a personal computer.
  • the user terminal is one selected from the group consisting of a smartphone, a tablet terminal, and a personal computer.
  • the disease image is an image taken with a smartphone camera or an image from an image library, focusing on the affected area.
  • the position of the position information is the user's current position or the user's specified position.
  • the disease name is provided as one or more disease candidate names.
  • the store is a pharmacy or drug store.
  • the product DB is a DB containing at least one selected from the group consisting of commercially available or homemade medical drugs, over-the-counter drugs, and functional foods including FOSHU and functional foods, [1]- The device according to any one of [9].
  • the disease diagnosis means is a device connected to the treatment diagnosis device via a network.
  • the device according to any one of [2] to [11], wherein the inventory management device is part of the treatment support device.
  • the inventory control device is an inventory control device for one store or an inventory control device for a plurality of stores.
  • the control means further comprises a user DB, The control means receives user registration from the user terminal through the transmission/reception means, records user registration information in a user DB of the control means, and returns a user ID and password to the user terminal. , [1] to [13].
  • control means receives a request for product ordering and/or shipping from the user terminal through the transmitting/receiving means, and transmits request information for product ordering and/or shipping to the store management device; ].
  • the control means further comprises a medical institution DB and is connected to a medical appointment device of the medical institution via a network;
  • a medical institution DB is inquired about a medical institution where a skin disease specialist treats based on the user information, and the medical institution information is acquired, and the user terminal
  • the device according to any one of [1] to [15], which transmits to [17]
  • the control means further acquires medical appointment information from the medical appointment device and transmits it to the user terminal,
  • the apparatus according to [16] which transmits medical treatment request information transmitted from the user terminal to the medical treatment reservation device, and transmits reception information transmitted from the medical treatment reservation device to the user terminal.
  • the disease diagnosis means comprises a data group of disease images labeled with disease names, a conversion unit that converts each disease image into feature amount data, and a neural network model subjected to machine learning processing on the data group,
  • the disease image to be diagnosed transmitted from the control means is converted into feature amount data by the conversion unit, calculated by the neural network model, converted into feature amount data, and the signs are identified to make a diagnosis including the name of the disease. obtaining information and sending it to the control means;
  • the device according to any one of [1] to [17].
  • the apparatus of [18] wherein the neural network model is a deep convolutional neural network model.
  • the data set comprises 50,000 to 400,000 disease images.
  • a treatment support system comprising a skin disease treatment support device and a user terminal connected via a network,
  • the treatment support device is User information including disease images and location information is received from the user terminal, and diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and the product are transmitted to the user terminal.
  • a system comprising: [27]
  • the treatment support device is further connected to a store management device via a network, the transmission means further transmits store information including inventory information, and the control means further controls the inventory of the product from the store.
  • a treatment support method using a system comprising a skin disease treatment support device and a user terminal connected via a network, a step of receiving user information including a disease image and location information from the user terminal by means of transmission/reception means in the treatment support apparatus; a step of acquiring diagnostic information from a disease diagnosis means based on the disease image transmitted from the transmission/reception means by the control means in the treatment support apparatus; obtaining a product name corresponding to the disease name in the diagnostic information from the product DB by means of control means in the treatment support device; Based on the location information of the user information, store information including the name and location of the store near the location and the store from the online store is acquired from the store DB by the control means of the treatment support device.
  • Diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and a store including a store name and a store location where the product can be obtained are sent to the user terminal by the transmitting/receiving means of the treatment support device. transmitting the information;
  • a method, including [29] The treatment support device is further connected to a store management device via a network, a step of acquiring inventory information including a store having inventory of the product and an inventory amount from the store management device by the control means of the treatment support device; a step of further transmitting store information including the stock information to the user terminal;
  • the method of [28] comprising
  • the patient himself/herself makes an auxiliary diagnosis of the skin disease, acquires product information such as pharmaceuticals or functional foods based on the diagnosis, and obtains the product information. Since it is possible to obtain information on stores that have stocks of these products and treat them themselves, it is possible to improve the convenience of self-treatment for users and contribute to the development of self-medication.
  • FIG. 4 is a diagram showing an outline of one form of an interface screen of the user terminal (200) according to the embodiment of the present invention
  • FIG. 4 is a diagram showing an outline of one form of an interface screen of the user terminal (200) according to the embodiment of the present invention.
  • BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the outline
  • BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the outline
  • 1 is a diagram showing an outline of one form of disease diagnosis means (130) in a treatment support device (100) according to an embodiment of the present invention;
  • FIG. It is a figure which shows the flow of the treatment assistance method which concerns on embodiment of this invention.
  • the treatment support device according to the present invention is the device (100) shown in FIG. is a treatment support device.
  • the treatment support apparatus (100) receives user information including a disease image and position information from the user terminal (200), and transmits diagnostic information including a disease name and the diagnosis to the user terminal (200).
  • the apparatus is characterized by comprising control means (120) for processing diagnostic information, the product information, and the store information according to the user information and transmitting the processed information to the transmitting/receiving means.
  • the medical treatment support device (100) further comprises a store management device (300) to which the medical care support device (100) is connected via a network (900).
  • the transmitting/receiving means (110) further transmits store information including inventory information
  • the control means (120) further manages the store having inventory of the product and the inventory information including the inventory amount from the store. It is characterized by being a device that acquires from a device (300).
  • a treatment support device (100) according to the present invention is connected to a user terminal (200) and a store management device (300) via a network (900) as shown in FIG.
  • the network (900) various networks can be cited as long as they are computer networks. Preferably, it is the Internet.
  • the treatment support device (100) can include a server inside it, and can also be connected to a server installed separately outside it. Also, the treatment support device (100) is preferably a so-called cloud computer.
  • the cloud computer allows the user using the user terminal (200) to use computer processing via the network (900).
  • the treatment support apparatus (100) acquires diagnostic information including a disease name, product information responding to the diagnostic information, and the product based on a disease image or the like transmitted by a user. It is possible to quickly provide users with store information for Users can easily obtain diagnostic information for their own skin diseases, product information for treatment assistance, and store information for product acquisition. As a result, it is possible for the user to perform supplementary treatment for skin diseases by himself/herself, contributing to self-medication.
  • the users of the user terminal (200) in the treatment support device (100) include patients, general consumers, and medical personnel.
  • patients use the treatment support apparatus (100) to assist treatment of their own skin diseases, general consumers to assist treatment of relatives and acquaintances, and medical personnel to assist treatment of patients. can be used.
  • a smart phone As the user terminal (200) in the treatment support device (100), a smart phone, a tablet PC, a personal computer, etc. can be mentioned. Among them, mobile terminals such as smart phones and tablet PCs are preferable from the viewpoint of convenience that they can be used anywhere.
  • the disease image transmitted from the user terminal (200) is preferably an image focused on the affected area.
  • Examples of the disease image include an image captured by a smartphone camera and an image from an image library.
  • the disease image can be uploaded to the treatment support device from the interface screen provided on the user terminal, for example, as shown in FIGS.
  • the number of disease images is preferably one or two.
  • One image is preferable from the point of convenience for the user, and two images are preferable from the point of sensitivity and specificity of diagnostic results.
  • When uploading two disease images it is preferable that one image is a close-up image of the affected area and the other image is an image of the entire disease.
  • the location information transmitted from the user terminal (200) is generally the user's current location or the user's specified location.
  • the designated position can be determined by inputting or selecting an address on the interface screen, such as the user's home or workplace, or by mapping it on a map.
  • the user information other than the location information of the user includes a user ID, a user password, selection information about the diagnostic information returned from the treatment support device (100) to the user terminal (200), and the like. can be mentioned.
  • the selection information includes the number of candidate disease names to be returned (e.g., 1 disease, 3 diseases, 5 diseases, etc.), the f-1 score of the diagnosis result, the presence or absence of display of sensitivity and specificity, Examples include the presence or absence of an ICD (International Classification of Diseases) code for a disease. You can also ask for comments from a dermatologist.
  • ICD International Classification of Diseases
  • FIG. 4 shows an example of the interface screen on which the user inputs the user information on the user terminal (200).
  • the user can upload two disease images.
  • the user's comments can also be entered if the user solicits the specialist's comments.
  • the user can also enter location information, either the current location or a specified location for product acquisition based on diagnostic results.
  • diagnostic information including a disease name and product information including a product name responding to the diagnostic information
  • store information including store name, store location and inventory information where the product can be obtained.
  • the product in the product information returned from the treatment support device (100) to the user terminal (200) is not particularly limited as long as it can be used as a treatment or treatment aid for skin diseases, and various products.
  • can give Among these, ethical drugs (hereinafter also referred to as "prescription drugs”), over-the-counter drugs (hereinafter also referred to as "OTC drugs”), functional foods such as FOSHU and functionally labeled foods are preferred.
  • Examples of the product information other than the product name include indications, efficacy/effects, health promoting action, functionality, and side effects of the product.
  • a pharmacy, a drug store, etc. can be mentioned as the store in the store information returned from the treatment support device (100) to the user terminal (200).
  • the store information returned from the treatment support device (100) to the user terminal (200) can be mentioned.
  • the store location in the store information can be displayed by entering an address or specifying a location on a map on the interface screen provided on the user terminal (200).
  • the inventory information in the store information is displayed on the interface screen provided to the user terminal (200) as the presence or absence of inventory, the inventory quantity, etc. together with the store information of the store that has the inventory of the product. can be done.
  • the information transmitted from the user terminal (200) to the transmitting/receiving means (110) in the treatment support device (100) is further transmitted to the control means (120).
  • the control means (120) can transmit the user information including the disease image to the disease diagnosis means (130) and acquire the diagnosis information including the disease name returned from the disease diagnosis means (130). can.
  • the control means (120) inquires the product DB (121) for product information corresponding to the disease name based on the diagnostic information, and stores the store DB (121) based on the product information and the location information of the user. 122) for store information including the store name to acquire the product information and the store information.
  • the control means (120) can further transmit the store information to the store management device (300) and receive a reply about the name of the store that has the product in stock and the stock amount.
  • the control means (120) processes the diagnostic information, the product information, and the store information based on the user information, and can reply to the user terminal (200) through the transmitting/receiving means (110). can.
  • FIG. 5 shows an example of the interface screen on which the user displays information transmitted from the treatment support device (100) on the user terminal (200).
  • the ICD codes of two candidate diseases, disease names, and specialist comments are displayed on the interface screen as the diagnosis information.
  • the product information the product classification, product name, and product number of two products for the disease 1 are displayed.
  • the store name, address, telephone number, and map display button are displayed as store information.
  • the disease diagnosis means (130) AI-processes the disease image transmitted from the control means (120) and returns diagnostic information including the disease name to the control means (120).
  • the disease diagnosis means (130) may be a part of the treatment support device (100), or may be a device connected to the treatment diagnosis device (100) via a network (900). In this case, the disease diagnosis means (130) may be a disease diagnosis device of an external organization.
  • a commercially available or homemade database can be used as the product DB (121).
  • the product DB (121) may be a database of ethical drugs, over-the-counter drugs, or functional foods including FOSHU and functionally labeled foods.
  • a commercially available database, a database provided by a pharmacy chain, or a homemade database can be used as the store DB (122). Further, the store DB can include the net store.
  • the inventory management device (300) is connected to the treatment diagnosis device (100) via a network, even if it is an inventory DB (not shown) which is a part of the treatment support device (100). It may be a device with
  • the inventory management device (300) may be an inventory management device for one store or, for example, an inventory management device that controls a plurality of stores under a chain.
  • a user who uses the user terminal (200) can perform user registration of the treatment support device (100) from the user terminal (200).
  • the control means (120) further comprises a user DB (123) as shown in FIG.
  • User registration is accepted, user registration information is recorded in the user DB of the control means (120), and the user ID and password are returned to the user terminal.
  • various methods such as one-time password, bar code, QR code, and biometric authentication can be used in addition to general ID and password.
  • the treatment support device (100) can assist in arranging the product for the user.
  • the treatment support device (100) receives the order reservation for the product and the delivery request for the product from the user terminal (200) through the transmitting/receiving means (110), and through the control means (120), the store management.
  • An order reservation for the product and a delivery request for the product can be sent to the device (300). Users can easily obtain products such as OTC medications for therapeutic assistance.
  • the store management device (300) is a device that manages inventory management of the store, order reservations for the products, and delivery requests for the products.
  • the device may be one device that performs each of the above functions, or may be a plurality of devices.
  • the treatment support device (100) can support the user's medical care at a medical institution and, in turn, support the acquisition of prescription drugs.
  • the treatment support device (100) further comprises a medical institution DB (124) and is connected to the medical institution's medical appointment device (400) via a network (900).
  • the treatment support device (100) inquires of a medical institution DB (124) about a medical institution where a dermatology specialist treats. , medical institution information can be acquired and transmitted to the user terminal (200).
  • the disease diagnosis means (130) in the treatment support device (100) includes a data group (131) of disease images labeled with disease names. , a conversion unit (132) for converting each disease image into feature amount data, and a neural network model (133) subjected to machine learning processing with the data group,
  • the disease image to be diagnosed transmitted from the control means (120) is converted into feature amount data by the conversion unit (132), calculated by the neural network model (133), converted into feature amount data, and classified. is identified, diagnostic information including the disease name is acquired, and the diagnostic information including the disease name is transmitted to the control means (120).
  • the neural network model subjected to machine learning processing on the data group in order to detect the feature amount is a deep convolutional neural network model suitable for image analysis.
  • the disease name can have tens to more than 100 classifications (hereinafter also referred to as “categories”) as indicators of individual data in the data group. For example, having categories such as about 50, about 75, and about 100.
  • 45 or 70 disease categories are selected as specific aspects of the disease name and implemented in the disease diagnosis means.
  • the data set includes 50,000 to 500,000 or more labeled disease images. These disease images have labels (disease names) determined by a doctor specializing in skin diseases.
  • the labeled data group for each indicator has a sufficient number of disease images for determining the diagnosis result.
  • the feature amount refers to the amount that characterizes the disease image.
  • this feature amount data can be represented by numerical values such as multidimensional vectors.
  • Each of the disease images can be converted into the feature data by an algorithm, also called feature extractor. Examples of such algorithms include SIFT (Scale-Invariant Feature Transform) and HOG (Histograms of Oriented Gradients).
  • the deep convolutional neural network model first creates a hypothetical deep neural network model, trains the deep neural network model with a large amount of disease image data group, calculates the feature amount, and corrects the feature amount. Comparing with the feature quantity, correcting the calculation formula, and repeating these steps completes the learning, and the target deep convolutional neural network model can be obtained.
  • the deep convolutional neural network model refers to a deep neural network in which the layers of the neural network are deepened many times, and which is defined by dividing the layers of the neural network into a convolutional layer and a pooling layer. .
  • the deep convolutional neural network model is preferably a multilayer, for example, 34-layer ResNet (Residual Network) model.
  • the disease name can be provided as multiple disease name candidates. For example, 2, 3, 5, or 10 disease candidate names can be sent to the control means (120).
  • the diagnostic information can include the f1-score, sensitivity, and specificity of the disease (class). For example, if the disease name is 5, information such as sensitivity >75% and specificity >98% can be included.
  • the percentage display does not indicate the probability of the disease name, but represents the percentage based on the diagnostic defined image. Sensitivity is the rate of correctly determined to be in the corresponding class, and specificity can be said to be the rate of correctly determined not to be in the relevant class.
  • the f1-score represents the average of precision and recall.
  • the disease diagnosis means (130) will be further explained by First Derm (registered trademark) as a specific embodiment of the disease diagnosis means (130).
  • FirstDerm registered trademark
  • pytorch a deep convolutional neural network
  • markers classifying skin diseases into 45 and 70 categories are used to create a tentative model using the Fasti library using the pie torch.
  • the model of the deep convolutional neural network is based on a 34-layer ResNet fine-tuned to the image that the user sent to the treatment support device (100).
  • the model uses one-cycle super-convergence for fast learning time, as well as discriminative layer learning to ensure optimal transfer learning.
  • Transfer Learning is a method of making learning (learned model) in one area useful in another area for efficient learning.
  • the model uses a simultaneous input method to capture metadata and text input. To ensure that each final label contains a sufficient number of example images, we use a label hierarchy with multiple layers of meaningful coarse labels. Oversampling and thresholding are performed to handle class imbalance. The final accuracy is improved by a homegrown TTA technique (test-time augmentation procedure) and ensemble learning.
  • Diseases as markers of the 45 classes are acne, actinic keratosis, hemangiomas, atopic dermatitis, atypical moles, balanitis, basal cell carcinoma, Lyme disease, frostbite, genital warts, and contact dermatitis.
  • the treatment support system according to the present invention includes a skin disease treatment support device (100) connected via a network (900), a user terminal ( 200), and optionally a store management device (300), wherein the treatment support device receives user information including disease images and location information from the user terminal (200),
  • the user terminal (200) includes diagnostic information including disease name, product information including product name responding to the diagnostic information, store name and store location where the product can be obtained, and arbitrarily inventory information.
  • the system is characterized by comprising a control means (120) for processing the diagnostic information, the product information, and the store information according to the user information and transmitting the processed information to the transmitting/receiving means (110).
  • a step (S2) of acquiring the diagnostic information from the disease diagnostic means (130) and a step of acquiring the product name corresponding to the disease name of the diagnostic information from the product DB (121) by the control means in the treatment support device ( S3), based on the location information of the user information, store information including store names and store locations of stores near the location and stores from the Internet store is stored in the store DB by the control means in the treatment support device.
  • the control means in the treatment support device transfers the inventory information including stores having inventory of the product from among the stores and the inventory amount to the store management device (300). ), and the transmitting/receiving means (110) in the treatment support device (100), the diagnostic information including the disease name and the product name responding to the diagnostic information are sent to the user terminal (200) and a step (S7) of transmitting store information including the store name, store location, and inventory information where the product can be obtained.
  • Each functional processing unit reads a computer program stored in a storage device such as a ROM (Read Only Memory) or a hard disk by a CPU (Central Processing Unit) built into the computer device, and reads the computer program executed by the CPU.
  • a storage device such as a ROM (Read Only Memory) or a hard disk by a CPU (Central Processing Unit) built into the computer device, and reads the computer program executed by the CPU.
  • each functional processing unit reads and writes necessary data such as a table from a database stored in a storage device and a storage area on a memory, and depending on the case, related hardware (for example, input output device, display device, communication interface device).
  • the database (DB) in the embodiment of the present invention may be a commercial database, but it also means a mere collection of tables and files, and the internal structure of the database itself does not matter.
  • the database may also be considered as one server (database server).
  • treatment support device 110 transmission/reception means 120 control means 121 product DB 122 store database 123 User DB 124 Medical institution database 130 disease diagnosis means 131 data group 132 conversion unit 133 neural network 200 user terminal 300 store management device 400 medical appointment device

Abstract

[Problem] The problem addressed by the present invention is to provide a treatment assistance device, system, and method that contribute to self-medication and utilize AI diagnostic imaging. [Solution] This device is a treatment assistance device for skin diseases that is connected to a user terminal via a network, wherein the device is constituted by: a transmission/reception means that receives user information from the user terminal and transmits diagnosis information, product information, and store information to the user terminal; and a control means that acquires the diagnosis information from a disease diagnosis means on the basis of disease images, acquires a product name from a product database, acquires the store information from a store database on the basis of position information in the user information, discretionarily acquires stock information from a store management device, processes the diagnosis information, the product information, and the store information responsive to the disease images in accordance with the user information, and transmits the processed information to the transmission/reception means.

Description

治療支援装置、システム、及び方法TREATMENT SUPPORT DEVICE, SYSTEM, AND METHOD
　本発明は、皮膚疾患の治療支援装置、システム、及び方法に関し、特に、セルフメディケーションに貢献するＡＩ画像診断を利用した治療支援装置、方法、及びシステムに関する。 The present invention relates to a treatment support device, system, and method for skin diseases, and more particularly to a treatment support device, method, and system using AI image diagnosis that contributes to self-medication.
　皮膚疾患の分野において専門医師は比較的少なく、専門医師による患者の診察機会は限定的である。このため、患者は往々にして自分自身で診断情報を得ようとすることがある。しかし、この場合には診断結果の間違いも多く、皮膚疾患診断の問題点と指摘されている。 There are relatively few specialists in the field of skin diseases, and opportunities for specialists to examine patients are limited. For this reason, patients often seek diagnostic information on their own. However, in this case, there are many errors in the diagnosis results, which has been pointed out as a problem in diagnosing skin diseases.
　また、皮膚疾患の専門医師でも正確な診断は容易ではなく、専門医師ではない医療関係者にとって皮膚疾患の診断は困難をともなう。このため、皮膚疾患の診断のための補助的ツールが期待されている。 In addition, accurate diagnosis is not easy even for skin disease specialists, and skin disease diagnosis is difficult for medical professionals who are not specialists. Therefore, an auxiliary tool for the diagnosis of skin diseases is expected.
　近年、ＩＴ技術、特に最近は人工知能（以下、「ＡＩ」ともいう。）技術の発達は目覚ましく、これらの技術を利用した診断に期待がもたれている。例えば、医師と医師をつなぐDoctor-to-Doctor（ＤｔｏＤ）型皮膚科遠隔診療サービス、アプリを介した医師による有料の遠隔医療診断などがその例として挙げられる。しかし、これらは患者や一般需要者が利用可能なアプリではなく、医師同士又はＢｔｏＢを前提としたものが多い。 In recent years, the development of IT technology, especially artificial intelligence (hereinafter also referred to as "AI") technology has been remarkable recently, and expectations are high for diagnosis using these technologies. Examples include a Doctor-to-Doctor (DtoD) dermatology telemedicine service that connects doctors and paid telemedicine diagnosis by a doctor via an app. However, many of these applications are not applications that can be used by patients or general consumers, but are based on the premise of doctor-to-doctor or BtoB.
　皮膚疾患の診断は画像の解析の側面を有している。医療画像の解析用のシステムとして人工知能を利用した発明は知られているが、皮膚疾患の診断を対象とする発明ではない（特許文献１を参照）。 Diagnosis of skin diseases has aspects of image analysis. Inventions using artificial intelligence as systems for analyzing medical images are known, but they are not inventions intended for diagnosing skin diseases (see Patent Document 1).
　一方、セルフメディケーションのツールとして、患者自身の治療行為を支援や補助する、患者や一般需要者が利用することのできるＡＩ技術の活用は課題とされている。 On the other hand, as a tool for self-medication, the utilization of AI technology that can be used by patients and general consumers to support and assist the patient's own treatment is an issue.
特開２０１９－５３６１３２号公報JP 2019-536132 A
　皮膚疾患のセルフメディケーションを進展させるためには、患者自身が何らかの診断結果を取得して、これを治療等する医薬品等の製品を確保して、自己治療を行うことが課題となる。 In order to advance self-medication for skin diseases, it is a challenge for patients themselves to obtain some kind of diagnostic results, secure products such as medicines to treat them, and carry out self-treatment.
　すなわち、本発明は、セルフメディケーションに貢献するＡＩ画像診断を利用した治療支援装置、システム、及び方法を提供することを課題とする。 That is, an object of the present invention is to provide a treatment support device, system, and method using AI image diagnosis that contributes to self-medication.
　すなわち、本発明は以下のとおりである。
［１］
　利用者端末とネットワークを介して接続された皮膚疾患の治療支援装置であって、当該治療支援装置が、
　前記利用者端末からの疾患画像及び位置情報を含む利用者情報を受信し、前記利用者端末に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名及び店舗場所を含む店舗情報を送信する送受信手段、
　前記送受信手段から送信される前記疾患画像に基づいて、疾患診断手段から前記診断情報を取得し、
　製品ＤＢから前記診断情報の疾患名に対応する前記製品名を取得し、
　前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢから取得し、及び
　前記疾患画像に応答する前記診断情報、前記製品情報、及び前記店舗情報を前記利用者情報に応じて処理して、前記送受信手段に送信する制御手段、
を備える装置。
［２］
　前記治療支援装置が、さらにネットワークを介して店舗管理装置と接続され、前記送受信手段が、さらに在庫情報を含む店舗情報を送信し、及び
　前記制御手段が、さらに前記店舗の中から前記製品の在庫を有する店舗及び在庫量を含む在庫情報を店舗管理装置から取得する、［１］に記載の装置。
［３］
　前記利用者端末の利用者が、患者、一般需要者、医療関係者からなる群から選択されるいずれかの利用者である、［１］又は［２］に記載の装置。
［４］
　前記利用者端末が、スマートフォン、タブレット端末、パソコンからなる群から選択されるいずれかである、［１］～［３］のいずれか１項に記載の装置。
［５］
　前記疾患画像が、患部に焦点をあてた、スマートフォンカメラ撮影による画像又は画像ライブラリからの画像である、［１］～［４］のいずれか１項に記載の装置。
［６］
　前記位置情報の位置が、前記利用者の現在位置又は前記利用者の指定位置である、［１］～［５］のいずれか１項に記載の装置。
［７］
　前記疾患名が１又は複数の疾患候補名で提供される、［１］～［６］のいずれか１項に記載の装置。
［８］
　前記製品が、医療用医薬品、一般用医薬品、並びにトクホ及び機能性表示食品を含む機能性食品からなる群から選択される、［１］～［７］のいずれか１項に記載の装置。
［９］
　前記店舗が、薬局又はドラッグストアである、［１］～［８］のいずれか１項に記載の装置。
［１０］
　前記製品ＤＢが、市販又は自家製の、医療用医薬品、一般用医薬品、並びにトクホ及び機能性表示食品を含む機能性食品からなる群から選択される少なくとも1つを含むＤＢである、［１］～［９］のいずれか１項に記載の装置。
［１１］
　前記疾患診断手段が、前記治療診断装置とネットワークを介して接続されている装置である、［１］～［１０］のいずれか１項に記載の装置。
［１２］
　前記在庫管理装置が、前記治療支援装置の一部分である、［２］～［１１］のいずれか１項に記載の装置。
［１３］
　前記在庫管理装置が、一店舗の在庫管理装置又は複数の店舗の在庫管理装置である、［２］～［１２］のいずれか１項に記載の装置。
［１４］
　前記制御手段が、さらに利用者ＤＢを備え、
　前記制御手段が前記送受信手段を通して、前記利用者端末からの利用者登録を受けつけ、利用者登録情報を前記制御手段の利用者ＤＢに記録し、前記利用者端末に利用者ＩＤ及びパスワードを返信する、［１］～［１３］のいずれか１項に記載の装置。
［１５］
　前記制御手段が前記送受信手段を通して、前記利用者端末からの製品発注及び／又は発送の依頼を受けつけ、前記店舗管理装置に製品発注及び／又は発送の依頼情報を送信する、［１］～［１４］のいずれか１項に記載の装置。
［１６］
　前記制御手段が、さらに医療機関ＤＢを備え、医療機関の診療予約装置とネットワークを介して接続され、
　前記製品情報における製品が医療用医薬品であるとき、前記利用者情報に基づいて皮膚疾患の専門医師が診療する医療機関を医療機関ＤＢに問合せして、医療機関情報を取得し、前記利用者端末に送信する、［１］～［１５］のいずれか１項に記載の装置。
［１７］
　前記制御手段が、さらに前記診療予約装置から診療予約情報を取得して、前記利用者端末に送信し、
　前記利用者端末から送信される診療希望情報を前記診療予約装置に送信して、前記診療予約装置から送信される受付情報を前記利用者端末に送信する、［１６］に記載の装置。
［１８］
　前記疾患診断手段が、疾患名の標識を付された疾患画像のデータ群、各疾患画像を特徴量データに変換する変換部、前記データ群で機械学習処理が施されたニューラルネットワークモデルを備え、
　前記制御手段から送信された診断対象の疾患画像を前記変換部で特徴量データに変換し、前記ニューラルネットワークモデルにより計算して特徴量データに変換し、前記標識を識別して疾患名を含む診断情報を取得し、前記制御手段に送信する、
［１］～［１７］のいずれか１項に記載の装置。
［１９］
　前記ニューラルネットワークモデルが、深層畳み込みニューラルネットワークモデルである、［１８］に記載の装置。
［２０］
　前記データ群が、５０，０００から４００，０００の疾患画像を含む、［１８］又は［１９］に記載の装置。
［２１］
　前記深層畳み込みニューラルネットワークが、３４層のＲｅｓＮｅｔ（Residual Network）モデルである、［１９］又は［２０］に記載の装置。
［２２］
　前記疾患名が、ＩＣＤコードの疾患カテゴリーで表される、［１８］～［２１］のいずれか１項に記載の装置。
［２３］
　前記疾患名が、４５又は７０の疾患カテゴリーで表される、［１８］～［２２］のいずれか１項に記載の装置。
［２４］
　前記診断情報の疾患名が、１～５種のいずれか１つの疾患カテゴリーの疾患候補名として提示される、［１８］～［２３］のいずれか１項に記載の装置。
［２５］
　前記診断情報が、ｆ１－スコア、感度、及び特異性の数値を含む、［１８］～［２４］のいずれか１項に記載の装置。
［２６］
　ネットワークを介して接続された皮膚疾患の治療支援装置及び利用者端末を備える治療支援システムであって、
　前記治療支援装置が、
　前記利用者端末からの疾患画像及び位置情報を含む利用者情報を受信し、前記利用者端末に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名及び店舗場所を含む店舗情報を送信する送受信手段、
　前記送受信手段から送信される前記疾患画像に基づいて、疾患診断手段から前記診断情報を取得し、
　製品ＤＢから前記診断情報の疾患名に対応する前記製品名を取得し、
　前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢから取得し、及び
　前記疾患画像に応答する前記診断情報、前記製品情報、及び前記店舗情報を前記利用者情報に応じて処理して、前記送受信手段に送信する制御手段、
を備える、システム。
［２７］
　前記治療支援装置が、さらにネットワークを介して店舗管理装置と接続され、前記送信手段が、さらに在庫情報を含む店舗情報を送信し、及び
　前記制御手段が、さらに前記店舗の中から前記製品の在庫を有する店舗及び在庫量を含む在庫情報を店舗管理装置から取得する、［２６］に記載のシステム。
［２８］
　ネットワークを介して接続された皮膚疾患の治療支援装置及び利用者端末を備えるシステムを用いる治療支援方法であって、
　前記治療支援装置における送受信手段により、前記利用者端末からの疾患画像及び位置情報を含む利用者情報を受信する工程、
　前記治療支援装置における制御手段により、前記送受信手段から送信される前記疾患画像に基づいて、疾患診断手段から診断情報を取得する工程、
　前記治療支援装置における制御手段により、製品ＤＢから前記診断情報の疾患名に対応する製品名を取得する工程、
　前記治療支援装置における制御手段により、前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢから取得する工程、
　前記治療支援装置における前記送受信手段により、前記利用者端末に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名及び店舗場所を含む店舗情報を送信する工程、
を含む、方法。
［２９］
　前記治療支援装置が、ネットワークを介して、さらに店舗管理装置と接続され、
　前記治療支援装置における制御手段により、さらに前記店舗の中から前記製品の在庫を有する店舗及び在庫量を含む在庫情報を前記店舗管理装置から取得する工程、及び
　前記治療支援装置における前記送受信手段により、前記利用者端末に、さらに前記在庫情報を含む店舗情報を送信する工程、
を含む、［２８］に記載の方法。 That is, the present invention is as follows.
[1]
A skin disease treatment support device connected to a user terminal via a network, wherein the treatment support device
User information including disease images and location information is received from the user terminal, and diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and the product are transmitted to the user terminal. Sending/receiving means for transmitting store information including the store name and store location that can be acquired,
acquiring the diagnostic information from the disease diagnosis means based on the disease image transmitted from the transmission/reception means;
obtaining the product name corresponding to the disease name of the diagnostic information from the product DB;
Based on the location information of the user information, store information including store names and store locations of stores near the location and stores from the Internet store is acquired from the store DB, and responding to the disease image. control means for processing the diagnostic information, the product information, and the store information according to the user information and transmitting the information to the transmitting/receiving means;
A device comprising
[2]
The treatment support device is further connected to a store management device via a network, the transmitting/receiving means further transmits store information including inventory information, and the control means further controls the inventory of the product from the store. The device according to [1], which acquires inventory information including a store having a store and an inventory amount from a store management device.
[3]
The device according to [1] or [2], wherein the user of the user terminal is any user selected from the group consisting of patients, general consumers, and medical personnel.
[4]
The device according to any one of [1] to [3], wherein the user terminal is one selected from the group consisting of a smartphone, a tablet terminal, and a personal computer.
[5]
The apparatus according to any one of [1] to [4], wherein the disease image is an image taken with a smartphone camera or an image from an image library, focusing on the affected area.
[6]
The device according to any one of [1] to [5], wherein the position of the position information is the user's current position or the user's specified position.
[7]
The device according to any one of [1] to [6], wherein the disease name is provided as one or more disease candidate names.
[8]
The device according to any one of [1] to [7], wherein the product is selected from the group consisting of ethical drugs, over-the-counter drugs, and functional foods including FOSHU and functionally labeled foods.
[9]
The device according to any one of [1] to [8], wherein the store is a pharmacy or drug store.
[10]
The product DB is a DB containing at least one selected from the group consisting of commercially available or homemade medical drugs, over-the-counter drugs, and functional foods including FOSHU and functional foods, [1]- The device according to any one of [9].
[11]
The device according to any one of [1] to [10], wherein the disease diagnosis means is a device connected to the treatment diagnosis device via a network.
[12]
The device according to any one of [2] to [11], wherein the inventory management device is part of the treatment support device.
[13]
The device according to any one of [2] to [12], wherein the inventory control device is an inventory control device for one store or an inventory control device for a plurality of stores.
[14]
The control means further comprises a user DB,
The control means receives user registration from the user terminal through the transmission/reception means, records user registration information in a user DB of the control means, and returns a user ID and password to the user terminal. , [1] to [13].
[15]
[1] to [14], wherein the control means receives a request for product ordering and/or shipping from the user terminal through the transmitting/receiving means, and transmits request information for product ordering and/or shipping to the store management device; ].
[16]
the control means further comprises a medical institution DB and is connected to a medical appointment device of the medical institution via a network;
When the product in the product information is an ethical drug, a medical institution DB is inquired about a medical institution where a skin disease specialist treats based on the user information, and the medical institution information is acquired, and the user terminal The device according to any one of [1] to [15], which transmits to
[17]
The control means further acquires medical appointment information from the medical appointment device and transmits it to the user terminal,
The apparatus according to [16], which transmits medical treatment request information transmitted from the user terminal to the medical treatment reservation device, and transmits reception information transmitted from the medical treatment reservation device to the user terminal.
[18]
The disease diagnosis means comprises a data group of disease images labeled with disease names, a conversion unit that converts each disease image into feature amount data, and a neural network model subjected to machine learning processing on the data group,
The disease image to be diagnosed transmitted from the control means is converted into feature amount data by the conversion unit, calculated by the neural network model, converted into feature amount data, and the signs are identified to make a diagnosis including the name of the disease. obtaining information and sending it to the control means;
The device according to any one of [1] to [17].
[19]
The apparatus of [18], wherein the neural network model is a deep convolutional neural network model.
[20]
The apparatus of [18] or [19], wherein the data set comprises 50,000 to 400,000 disease images.
[21]
The apparatus according to [19] or [20], wherein the deep convolutional neural network is a 34-layer ResNet (Residual Network) model.
[22]
The device according to any one of [18] to [21], wherein the disease name is represented by an ICD code disease category.
[23]
The device according to any one of [18] to [22], wherein the disease name is represented by 45 or 70 disease categories.
[24]
The device according to any one of [18] to [23], wherein the disease name in the diagnostic information is presented as a disease candidate name in any one of 1 to 5 disease categories.
[25]
The apparatus of any one of [18]-[24], wherein the diagnostic information includes f1-score, sensitivity and specificity values.
[26]
A treatment support system comprising a skin disease treatment support device and a user terminal connected via a network,
The treatment support device is
User information including disease images and location information is received from the user terminal, and diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and the product are transmitted to the user terminal. Sending/receiving means for transmitting store information including the store name and store location that can be acquired,
acquiring the diagnostic information from the disease diagnosis means based on the disease image transmitted from the transmission/reception means;
obtaining the product name corresponding to the disease name of the diagnostic information from the product DB;
Based on the location information of the user information, store information including store names and store locations of stores near the location and stores from the Internet store is acquired from the store DB, and responding to the disease image. control means for processing the diagnostic information, the product information, and the store information according to the user information and transmitting the information to the transmitting/receiving means;
A system comprising:
[27]
The treatment support device is further connected to a store management device via a network, the transmission means further transmits store information including inventory information, and the control means further controls the inventory of the product from the store. The system according to [26], which acquires inventory information including a store having a store and an inventory amount from a store management device.
[28]
A treatment support method using a system comprising a skin disease treatment support device and a user terminal connected via a network,
a step of receiving user information including a disease image and location information from the user terminal by means of transmission/reception means in the treatment support apparatus;
a step of acquiring diagnostic information from a disease diagnosis means based on the disease image transmitted from the transmission/reception means by the control means in the treatment support apparatus;
obtaining a product name corresponding to the disease name in the diagnostic information from the product DB by means of control means in the treatment support device;
Based on the location information of the user information, store information including the name and location of the store near the location and the store from the online store is acquired from the store DB by the control means of the treatment support device. process,
Diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and a store including a store name and a store location where the product can be obtained are sent to the user terminal by the transmitting/receiving means of the treatment support device. transmitting the information;
A method, including
[29]
The treatment support device is further connected to a store management device via a network,
a step of acquiring inventory information including a store having inventory of the product and an inventory amount from the store management device by the control means of the treatment support device; a step of further transmitting store information including the stock information to the user terminal;
The method of [28], comprising
　本発明に係る皮膚疾患の治療支援装置、システム、及び方法によれば、患者自身が皮膚疾患の補助診断を行い、当該診断に基づいて医薬品又は機能性食品などの製品情報を取得し、当該製品の在庫を有する店舗情報を取得して、自ら治療を行うことができるため、利用者の自己治療における利便性を高め、セルフメディケーションの進展に寄与することができる。 According to the skin disease treatment support device, system, and method according to the present invention, the patient himself/herself makes an auxiliary diagnosis of the skin disease, acquires product information such as pharmaceuticals or functional foods based on the diagnosis, and obtains the product information. Since it is possible to obtain information on stores that have stocks of these products and treat them themselves, it is possible to improve the convenience of self-treatment for users and contribute to the development of self-medication.
本発明の実施形態に係る、治療支援システムの一形態の概要を示す図である。BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the outline|summary of one form of the treatment assistance system based on embodiment of this invention. 本発明の実施形態に係る、治療支援装置（１００）の一形態の概要を示す図である。BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the outline|summary of one form of a treatment assistance apparatus (100) based on embodiment of this invention. 本発明の実施形態に係る、治療支援装置（１００）の一形態の概要を示す図である。BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the outline|summary of one form of a treatment assistance apparatus (100) based on embodiment of this invention. 本発明の実施形態に係る、利用者端末（２００）のインターフェース画面の一形態の概要を示す図である。FIG. 4 is a diagram showing an outline of one form of an interface screen of the user terminal (200) according to the embodiment of the present invention; 本発明の実施形態に係る、利用者端末（２００）のインターフェース画面の一形態の概要を示す図である。FIG. 4 is a diagram showing an outline of one form of an interface screen of the user terminal (200) according to the embodiment of the present invention; 本発明の実施形態に係る、治療支援装置（１００）の一形態の概要を示す図である。BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the outline|summary of one form of a treatment assistance apparatus (100) based on embodiment of this invention. 本発明の実施形態に係る、治療支援装置（１００）の一形態の概要を示す図である。BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the outline|summary of one form of a treatment assistance apparatus (100) based on embodiment of this invention. 本発明の実施形態に係る、治療支援装置（１００）のにおける疾患診断手段（１３０）の一形態の概要を示す図である。1 is a diagram showing an outline of one form of disease diagnosis means (130) in a treatment support device (100) according to an embodiment of the present invention; FIG. 本発明の実施形態に係る治療支援方法のフローを示す図である。It is a figure which shows the flow of the treatment assistance method which concerns on embodiment of this invention.
　以下に、本発明の実施形態を図面に基づいて詳細に説明する。 Hereinafter, embodiments of the present invention will be described in detail based on the drawings.
（１）本発明に係る治療支援装置
　本発明に係る治療支援装置は、図１において示される装置（１００）であり、利用者端末（２００）とネットワーク（９００）を介して接続された皮膚疾患の治療支援装置である。当該治療支援装置（１００）は、前記利用者端末（２００）からの疾患画像及び位置情報を含む利用者情報を受信し、前記利用者端末（２００）に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗の店舗名及び店舗場所を含む店舗情報を送信する送受信手段（１１０）、前記送受信手段（１１０）から送信される前記疾患画像に基づいて、疾患診断手段（１３０）から前記診断情報を取得し、製品データベース（以下、「ＤＢ」ともいう。）（１２１）から前記疾患名に対応する前記製品名を取得し、前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢ（１２２）から取得し、及び前記疾患画像に応答する前記診断情報、前記製品情報、及び前記店舗情報を前記利用者情報に応じて処理して、前記送受信手段に送信する制御手段（１２０）、を備える装置であることを特徴とする。 (1) Treatment Support Device According to the Present Invention The treatment support device according to the present invention is the device (100) shown in FIG. is a treatment support device. The treatment support apparatus (100) receives user information including a disease image and position information from the user terminal (200), and transmits diagnostic information including a disease name and the diagnosis to the user terminal (200). Transmitting/receiving means (110) for transmitting product information including the product name in response to the information and store information including the name and location of the store where the product can be obtained, and the disease image transmitted from the transmitting/receiving means (110) acquires the diagnostic information from the disease diagnostic means (130), acquires the product name corresponding to the disease name from the product database (hereinafter also referred to as "DB") (121), and the user Based on the location information of the information, store information including store names and store locations of stores near the location and stores from the Internet store is acquired from the store DB (122), and the above-mentioned method for responding to the disease image is obtained. The apparatus is characterized by comprising control means (120) for processing diagnostic information, the product information, and the store information according to the user information and transmitting the processed information to the transmitting/receiving means.
　本発明に係る治療支援装置（１００）は、図２に示されるように、前記治療支援装置（１００）が、さらにネットワーク（９００）を介して接続される店舗管理装置（３００）を備え、前記送受信手段（１１０）が、さらに在庫情報を含む店舗情報を送信し、及び前記制御手段（１２０）が、さらに前記店舗の中から前記製品の在庫を有する店舗及び在庫量を含む在庫情報を店舗管理装置（３００）から取得する装置であることを特徴とする。 As shown in FIG. 2, the medical treatment support device (100) according to the present invention further comprises a store management device (300) to which the medical care support device (100) is connected via a network (900). The transmitting/receiving means (110) further transmits store information including inventory information, and the control means (120) further manages the store having inventory of the product and the inventory information including the inventory amount from the store. It is characterized by being a device that acquires from a device (300).
　本発明に係る治療支援装置（１００）は、図３のように、利用者端末（２００）及び店舗管理装置（３００）とネットワーク（９００）を介して接続されている。前記ネットワーク（９００）としては、コンピュータネットワークであれば種々のネットワークを挙げることができる。好ましくは、インターネットである。 A treatment support device (100) according to the present invention is connected to a user terminal (200) and a store management device (300) via a network (900) as shown in FIG. As the network (900), various networks can be cited as long as they are computer networks. Preferably, it is the Internet.
　前記治療支援装置（１００）は、その内部にサーバーを含むことができ、また、その外部に別個に設置されるサーバーと連結することもできる。また、前記治療支援装置（１００）は、いわゆるクラウドコンピュータであることが好ましい。クラウドコンピュータにより、利用者端末（２００）を使用する利用者は、コンピュータによる処理をネットワーク（９００）経由で利用することができる。 The treatment support device (100) can include a server inside it, and can also be connected to a server installed separately outside it. Also, the treatment support device (100) is preferably a so-called cloud computer. The cloud computer allows the user using the user terminal (200) to use computer processing via the network (900).
　前記治療支援装置（１００）は、以上のような構成により、利用者が送信した疾患画像等に基づいて、疾患名を含む診断情報、前記診断情報に応答する製品情報、及び前記製品を取得するための店舗情報を、利用者に迅速に提供できる。利用者は自己の皮膚疾患の診断情報、治療補助のための製品情報、製品取得のための店舗情報を容易に得ることができる。これにより、利用者自身による皮膚疾患の補助治療が可能となり、セルフメディケーションに貢献することができる。 With the configuration described above, the treatment support apparatus (100) acquires diagnostic information including a disease name, product information responding to the diagnostic information, and the product based on a disease image or the like transmitted by a user. It is possible to quickly provide users with store information for Users can easily obtain diagnostic information for their own skin diseases, product information for treatment assistance, and store information for product acquisition. As a result, it is possible for the user to perform supplementary treatment for skin diseases by himself/herself, contributing to self-medication.
　前記利用者端末（２００）においては、前記治療支援装置（１００）における送受信手段（１１０）により、前記利用者端末（２００）に提供される、例えば、図４及び５に示されるようなインターフェース画面を通して、前記利用者は前記治療支援装置（１００）との間で情報の送受信をすることができる。 In the user terminal (200), an interface screen as shown in FIGS. 4 and 5, for example, provided to the user terminal (200) by the transmitting/receiving means (110) in the treatment support device (100) Through, the user can transmit and receive information to and from the treatment support device (100).
　前記治療支援装置（１００）における利用者端末（２００）の利用者としては、患者、一般需要者、医療関係者などをあげることができる。一般に、患者は自身の皮膚疾患の治療補助のために、一般需要者は近親者や知人などの治療補助のために、医療関係者は患者治療の補助手段として、前記治療支援装置（１００）を使用することができる。 The users of the user terminal (200) in the treatment support device (100) include patients, general consumers, and medical personnel. In general, patients use the treatment support apparatus (100) to assist treatment of their own skin diseases, general consumers to assist treatment of relatives and acquaintances, and medical personnel to assist treatment of patients. can be used.
　前記治療支援装置（１００）における利用者端末（２００）としては、スマートフォン、タブレットＰＣ、パソコンなどをあげることができる。中でも、スマートフォン、タブレットＰＣのような携帯端末が、何処でも使用できる利便性の点から好ましい。 As the user terminal (200) in the treatment support device (100), a smart phone, a tablet PC, a personal computer, etc. can be mentioned. Among them, mobile terminals such as smart phones and tablet PCs are preferable from the viewpoint of convenience that they can be used anywhere.
　前記利用者端末（２００）から送信される疾患画像は、患部に焦点が当てられた画像が好ましい。前記疾患画像としては、スマートフォンカメラ撮影による画像、画像ライブラリからの画像をあげることができる。 The disease image transmitted from the user terminal (200) is preferably an image focused on the affected area. Examples of the disease image include an image captured by a smartphone camera and an image from an image library.
　前記疾患画像は、前記利用者端末に提供される、例えば、図４及び５に示されるような前記インターフェース画面から前記治療支援装置へアップロードすることができる。 The disease image can be uploaded to the treatment support device from the interface screen provided on the user terminal, for example, as shown in FIGS.
　前記疾患画像の数は、１画像又は２画像が好ましい。前記利用者の利便性の点からは１画像が好ましく、診断結果の感度及び特異性の点からは２画像が好ましい。２疾患画像をアップロードする場合は、１画像は患部の近接画像、他の１画像は疾患全体の画像であることが好ましい。 The number of disease images is preferably one or two. One image is preferable from the point of convenience for the user, and two images are preferable from the point of sensitivity and specificity of diagnostic results. When uploading two disease images, it is preferable that one image is a close-up image of the affected area and the other image is an image of the entire disease.
　前記利用者端末（２００）から送信される位置情報は、一般に、前記利用者の現在位置又は前記利用者の指定位置である。前記指定位置としては、前記利用者の自宅、職場などを挙前記インターフェース画面上で、住所を入力又は選択することによって、若しくは地図上にマッピングすることで、決定することができる。 The location information transmitted from the user terminal (200) is generally the user's current location or the user's specified location. The designated position can be determined by inputting or selecting an address on the interface screen, such as the user's home or workplace, or by mapping it on a map.
　前記利用者の位置情報以外の前記利用者情報としては、利用者ＩＤ、利用者パスワード、前記治療支援装置（１００）から前記利用者端末（２００）に返信される前記診断情報についての選択情報などを挙げることができる。 The user information other than the location information of the user includes a user ID, a user password, selection information about the diagnostic information returned from the treatment support device (100) to the user terminal (200), and the like. can be mentioned.
　前記選択情報としては、返信される候補となる疾患名の数（例えば、１疾患、３疾患、５疾患など）、診断結果についてのｆ－１スコア、感度、及び特異性についての表示の有無、疾患のＩＣＤ（国際疾病分類）コードの有無などをあげることができる。また、皮膚疾患の専門医のコメントを求めることもできる。 The selection information includes the number of candidate disease names to be returned (e.g., 1 disease, 3 diseases, 5 diseases, etc.), the f-1 score of the diagnosis result, the presence or absence of display of sensitivity and specificity, Examples include the presence or absence of an ICD (International Classification of Diseases) code for a disease. You can also ask for comments from a dermatologist.
　前記利用者端末（２００）において、利用者が前記利用者情報を入力する前記インターフェース画面の１例を図４に示す。この例において、利用者は２疾患画像をアップロードすることができる。利用者が専門医のコメントを求める場合には、利用者のコメントを入力することもできる。また、利用者は位置情報を、診断結果に基づく製品取得のために、現在位置又は指定位置を入力することができる。 FIG. 4 shows an example of the interface screen on which the user inputs the user information on the user terminal (200). In this example, the user can upload two disease images. The user's comments can also be entered if the user solicits the specialist's comments. The user can also enter location information, either the current location or a specified location for product acquisition based on diagnostic results.
　前記治療支援装置（１００）における送受信手段（１１０）により、前記利用者端末（２００）に提供される前記インターフェース画面を通して、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名、店舗場所及び在庫情報を含む店舗情報が前記送信される。 Through the interface screen provided to the user terminal (200) by the transmitting/receiving means (110) of the treatment support device (100), diagnostic information including a disease name and product information including a product name responding to the diagnostic information , and store information including store name, store location and inventory information where the product can be obtained.
　前記治療支援装置（１００）から前記利用者端末（２００）に返信される前記製品情報の製品としては、皮膚疾患の治療又は治療補助として使用できる製品であれば、特に限定されず、様々な製品をあげることができる。これらの中でも、医療用医薬品（以下、「処方箋薬」とも言う。）、一般用医薬品（以下、「ＯＴＣ薬」とも言う。）、トクホ及び機能性表示食品のような機能性食品が好ましい。 The product in the product information returned from the treatment support device (100) to the user terminal (200) is not particularly limited as long as it can be used as a treatment or treatment aid for skin diseases, and various products. can give Among these, ethical drugs (hereinafter also referred to as "prescription drugs"), over-the-counter drugs (hereinafter also referred to as "OTC drugs"), functional foods such as FOSHU and functionally labeled foods are preferred.
　前記製品名以外の前記製品情報としては、当該製品の適用症、効能・効果、健康増進作用、機能性、副作用などを挙げることができる。 Examples of the product information other than the product name include indications, efficacy/effects, health promoting action, functionality, and side effects of the product.
　前記治療支援装置（１００）から前記利用者端末（２００）に返信される前記店舗情報における店舗としては、薬局、ドラッグストアなどを挙げることができる。また、実在のある店舗ばかりでなく、ネットワーク上の仮想の店舗であるネット店舗を対象とすることもできる。 A pharmacy, a drug store, etc. can be mentioned as the store in the store information returned from the treatment support device (100) to the user terminal (200). In addition, not only existing stores but also online stores, which are virtual stores on the network, can be targeted.
　前記店舗情報における店舗場所は、前記利用者端末（２００）に提供される前記インターフェース画面上で、住所を入力、又は地図上の場所を指定するなどをして、表示することができる。 The store location in the store information can be displayed by entering an address or specifying a location on a map on the interface screen provided on the user terminal (200).
　前記店舗情報における在庫情報は、前記利用者端末（２００）に提供される前記インターフェース画面上で、前記製品の在庫を有する店舗の店舗情報と併せて、在庫の有無、在庫数量などとして表示することができる。 The inventory information in the store information is displayed on the interface screen provided to the user terminal (200) as the presence or absence of inventory, the inventory quantity, etc. together with the store information of the store that has the inventory of the product. can be done.
　前記治療支援装置（１００）における送受信手段（１１０）に、前記利用者端末（２００）から送信された情報は、さらに制御手段（１２０）に送信される。当該制御手段（１２０）は、疾患診断手段（１３０）に前記疾患画像を含む利用者情報を送信し、当該疾患診断手段（１３０）から返信された疾患名を含む前記診断情報を取得することができる。前記制御手段（１２０）は、当該診断情報に基づいて、前記製品ＤＢ（１２１）に疾患名に対応する製品情報を問合せし、さらに当該製品情報及び前記利用者の位置情報に基づいて店舗ＤＢ（１２２）に店舗名を含む店舗情報を問合せして、前記製品情報及び前記店舗情報を取得する。 The information transmitted from the user terminal (200) to the transmitting/receiving means (110) in the treatment support device (100) is further transmitted to the control means (120). The control means (120) can transmit the user information including the disease image to the disease diagnosis means (130) and acquire the diagnosis information including the disease name returned from the disease diagnosis means (130). can. The control means (120) inquires the product DB (121) for product information corresponding to the disease name based on the diagnostic information, and stores the store DB (121) based on the product information and the location information of the user. 122) for store information including the store name to acquire the product information and the store information.
　前記制御手段（１２０）は、さらに、前記店舗情報を店舗管理装置（３００）に送信して、前記製品の在庫を有する店舗の店舗名及び在庫量について返信を受けることができる。 The control means (120) can further transmit the store information to the store management device (300) and receive a reply about the name of the store that has the product in stock and the stock amount.
　前記制御手段（１２０）は、前記診断情報、前記製品情報、及び前記店舗情報を前記利用者情報に基づいて処理し、前記送受信手段（１１０）を通して前記利用者端末（２００）に返信することができる。 The control means (120) processes the diagnostic information, the product information, and the store information based on the user information, and can reply to the user terminal (200) through the transmitting/receiving means (110). can.
　前記利用者端末（２００）において、利用者は前記治療支援装置（１００）から送信される情報を表示する前記インターフェース画面の１例を図５に示す。この例において、前記インターフェース画面上には、前記診断情報として２候補疾患の前記ＩＣＤコード、疾患名、及び専門医コメントが表示される。前記製品情報として、前記疾患１について２製品の製品分類、製品名、及び製品番号が表示される。また、店舗情報として店舗名、住所、電話番号、及び地図表示ボタンが表示される。 FIG. 5 shows an example of the interface screen on which the user displays information transmitted from the treatment support device (100) on the user terminal (200). In this example, the ICD codes of two candidate diseases, disease names, and specialist comments are displayed on the interface screen as the diagnosis information. As the product information, the product classification, product name, and product number of two products for the disease 1 are displayed. In addition, the store name, address, telephone number, and map display button are displayed as store information.
　前記疾患診断手段（１３０）は、後述のとおり、前記制御手段（１２０）から送信される前記疾患画像をＡＩ処理して、疾患名を含む診断情報を前記制御手段（１２０）に返信する。 As will be described later, the disease diagnosis means (130) AI-processes the disease image transmitted from the control means (120) and returns diagnostic information including the disease name to the control means (120).
　前記疾患診断手段（１３０）は、前記治療支援装置（１００）の一部分であっても、前記治療診断装置（１００）とネットワーク（９００）を介して接続されている装置であってもよい。この場合、前記疾患診断手段（１３０）は、外部機関の疾患診断装置であってもよい。 The disease diagnosis means (130) may be a part of the treatment support device (100), or may be a device connected to the treatment diagnosis device (100) via a network (900). In this case, the disease diagnosis means (130) may be a disease diagnosis device of an external organization.
　前記製品ＤＢ（１２１）としては、市販又は自家製のデータベースを使用することができる。また、前記製品ＤＢ（１２１）は、医療用医薬品、一般用医薬品、又はトクホ及び機能性表示食品を含む機能性食品についてのデータベースであり得る。 A commercially available or homemade database can be used as the product DB (121). In addition, the product DB (121) may be a database of ethical drugs, over-the-counter drugs, or functional foods including FOSHU and functionally labeled foods.
　前記店舗ＤＢ（１２２）としては、市販、薬局チェーン提供又は自家製のデータベースを使用することができる。また、前記店舗ＤＢには、前記ネット店舗を含めることができる。 A commercially available database, a database provided by a pharmacy chain, or a homemade database can be used as the store DB (122). Further, the store DB can include the net store.
　前記在庫管理装置（３００）は、前記治療支援装置（１００）の一部分である在庫ＤＢ（図に示されていない）であっても、前記治療診断装置（１００）とネットワークを介して接続されている装置であってもよい。 The inventory management device (300) is connected to the treatment diagnosis device (100) via a network, even if it is an inventory DB (not shown) which is a part of the treatment support device (100). It may be a device with
　また、前記在庫管理装置（３００）は、一店舗の在庫管理装置であっても、例えば、チェーン傘下の複数の店舗を統括する在庫管理装置であってもよい。 In addition, the inventory management device (300) may be an inventory management device for one store or, for example, an inventory management device that controls a plurality of stores under a chain.
　前記利用者端末（２００）を利用する利用者は、前記利用者端末（２００）から前記治療支援装置（１００）の利用者登録をすることができる。 A user who uses the user terminal (200) can perform user registration of the treatment support device (100) from the user terminal (200).
　前記利用者登録として、前記制御手段（１２０）は、図６に示されるように、さらに利用者ＤＢ（１２３）を備え、前記送受信手段（１１０）を通して、前記利用者端末（２００）からの利用者登録を受けつけ、利用者登録情報を前記制御手段（１２０）の利用者ＤＢに記録し、前記利用者端末に利用者ＩＤ及びパスワードを返信する。利用者の認証は、一般のＩＤ及びパスワードの他にも、ワンタイム・パスワード、バーコード、ＱＲコード、生体認証など様々な方式を用いることができる。 As the user registration, the control means (120) further comprises a user DB (123) as shown in FIG. User registration is accepted, user registration information is recorded in the user DB of the control means (120), and the user ID and password are returned to the user terminal. For user authentication, various methods such as one-time password, bar code, QR code, and biometric authentication can be used in addition to general ID and password.
　前記治療支援装置（１００）は、前記利用者のために前記製品の手配を支援することができる。 The treatment support device (100) can assist in arranging the product for the user.
　前記治療支援装置（１００）は、前記送受信手段（１１０）を通して、利用者端末（２００）からの前記製品の注文予約及び前記製品の送付依頼を受けつけ、前記制御手段（１２０）を通して、前記店舗管理装置（３００）に前記製品の注文予約及び前記製品の送付依頼を送付することができる。利用者は、治療補助のＯＴＣ薬などの製品を容易に取得することができる。 The treatment support device (100) receives the order reservation for the product and the delivery request for the product from the user terminal (200) through the transmitting/receiving means (110), and through the control means (120), the store management. An order reservation for the product and a delivery request for the product can be sent to the device (300). Users can easily obtain products such as OTC medications for therapeutic assistance.
　ここで、前記店舗管理装置（３００）は、当該店舗の在庫管理及び前記製品の注文予約及び前記製品の送付依頼を管理する装置である。当該装置は、前記各機能を果たす一装置であってもよく、複数の装置であってもよい。 Here, the store management device (300) is a device that manages inventory management of the store, order reservations for the products, and delivery requests for the products. The device may be one device that performs each of the above functions, or may be a plurality of devices.
　前記治療支援装置（１００）は、前記利用者の医療機関における診療を支援し、引いては処方箋薬の取得を支援することができる。 The treatment support device (100) can support the user's medical care at a medical institution and, in turn, support the acquisition of prescription drugs.
　図７に示すように、前記治療支援装置（１００）は、さらに、医療機関ＤＢ（１２４）を備え、医療機関の診療予約装置（４００）とネットワーク（９００）を介して接続される。 As shown in FIG. 7, the treatment support device (100) further comprises a medical institution DB (124) and is connected to the medical institution's medical appointment device (400) via a network (900).
　前記治療支援装置（１００）は、前記製品情報における製品が医療用医薬品である場合、前記治療支援装置（１００）は皮膚疾患専門医師が診療する医療機関を医療機関ＤＢ（１２４）に問合せして、医療機関情報を取得し、利用者端末（２００）に送信することができる。 When the product in the product information is a medical drug, the treatment support device (100) inquires of a medical institution DB (124) about a medical institution where a dermatology specialist treats. , medical institution information can be acquired and transmitted to the user terminal (200).
　（２）前記治療支援装置における疾患診断手段
　図８に示されるように、前記治療支援装置（１００）における前記疾患診断手段（１３０）は、疾患名で標識された疾患画像のデータ群（１３１）、各疾患画像を特徴量データに変換する変換部（１３２）、前記データ群で機械学習処理が施されたニューラルネットワークモデル（１３３）を備え、
　前記制御手段（１２０）から送信された診断対象の疾患画像を前記変換部（１３２）で特徴量データに変換し、前記ニューラルネットワークモデル（１３３）により計算して特徴量データに変換して、分類を識別し、疾患名を含む診断情報を取得し、当該疾患名を含む診断情報を前記制御手段（１２０）に送信することを特徴とする。 (2) Disease diagnosis means in the treatment support device As shown in FIG. 8, the disease diagnosis means (130) in the treatment support device (100) includes a data group (131) of disease images labeled with disease names. , a conversion unit (132) for converting each disease image into feature amount data, and a neural network model (133) subjected to machine learning processing with the data group,
The disease image to be diagnosed transmitted from the control means (120) is converted into feature amount data by the conversion unit (132), calculated by the neural network model (133), converted into feature amount data, and classified. is identified, diagnostic information including the disease name is acquired, and the diagnostic information including the disease name is transmitted to the control means (120).
　前記特徴量を検出するために、前記データ群で機械学習処理が施されたニューラルネットワークモデルは、画像解析に適した深層畳み込みニューラルネットワークモデルであることが好ましい。 It is preferable that the neural network model subjected to machine learning processing on the data group in order to detect the feature amount is a deep convolutional neural network model suitable for image analysis.
　前記疾患名は、前記データ群の個々のデータの標識として、数１０から１００を超える分類（以下、「カテゴリー」ともいう。）を有することができる。例えば、約５０、約７５、及び約１００などのカテゴリーを有する。 The disease name can have tens to more than 100 classifications (hereinafter also referred to as "categories") as indicators of individual data in the data group. For example, having categories such as about 50, about 75, and about 100.
　前記疾患名の具体的な態様として、４５又は７０の疾患カテゴリーが選択され、前記疾患診断手段に実装されている。 45 or 70 disease categories are selected as specific aspects of the disease name and implemented in the disease diagnosis means.
　前記データ群は、標識が付された５０，０００から５００，０００、又はこれを越える数の疾患画像を含む。これらの疾患画像は、皮膚疾患の専門医師によって標識（疾患名）が確定されている。 The data set includes 50,000 to 500,000 or more labeled disease images. These disease images have labels (disease names) determined by a doctor specializing in skin diseases.
　前記標識の付された、各標識についてのデータ群は、診断結果の判定に十分な数の疾患画像数を有する。 The labeled data group for each indicator has a sufficient number of disease images for determining the diagnosis result.
　前記特徴量は、疾患画像を特徴づける量のことを言う。一般に、この特徴量のデータは多次元ベクトルのような数値で表すことができる。前記疾患画像のそれぞれを、特徴量抽出器とも呼ばれるアルゴリズムにより、前記特徴量データに変換することができる。当該アルゴリズムとしては、例えば、ＳＩＦＴ（Scale-Invariant Feature Transform）、ＨＯＧ（Histograms of Oriented Gradients）などを挙げることができる。 The feature amount refers to the amount that characterizes the disease image. In general, this feature amount data can be represented by numerical values such as multidimensional vectors. Each of the disease images can be converted into the feature data by an algorithm, also called feature extractor. Examples of such algorithms include SIFT (Scale-Invariant Feature Transform) and HOG (Histograms of Oriented Gradients).
　前記深層畳み込みニューラルネットワークモデルは、まず仮説としての深層ニューラルネットワークモデルを作成し、大量の前記疾患画像のデータ群で前記深層ニューラルネットワークモデルを学習させ、前記特徴量を計算し、当該特徴量を正しい特徴量と比較して計算式を補正し、これらの工程を繰り返すことにより学習を完了し、目的の深層畳み込みニューラルネットワークモデルとすることができる。 The deep convolutional neural network model first creates a hypothetical deep neural network model, trains the deep neural network model with a large amount of disease image data group, calculates the feature amount, and corrects the feature amount. Comparing with the feature quantity, correcting the calculation formula, and repeating these steps completes the learning, and the target deep convolutional neural network model can be obtained.
　前記深層畳み込みニューラルネットワークモデルは、ニューラルネットワークの層（layer）を何重にも深くした深層ニューラルネットワークについて、ニューラルネットワークの層を畳み込み層とプーリング層に分けて定義して実行するモデルのことをいう。 The deep convolutional neural network model refers to a deep neural network in which the layers of the neural network are deepened many times, and which is defined by dividing the layers of the neural network into a convolutional layer and a pooling layer. .
　前記深層畳み込みニューラルネットワークモデルについては、多層、例えば、３４層のＲｅｓＮｅｔ（Residual Network）モデルであることが好ましい。 The deep convolutional neural network model is preferably a multilayer, for example, 34-layer ResNet (Residual Network) model.
　前記疾患名は、複数の疾患名候補名として提供することができ。例えば、２、３、５、又は１０の疾患候補名として前記制御手段（１２０）に送信することができる。 The disease name can be provided as multiple disease name candidates. For example, 2, 3, 5, or 10 disease candidate names can be sent to the control means (120).
　前記診断情報は、前記疾患（クラス）のｆ１－スコア、感度、及び特異性を含むことができる。例えば、疾患名が５の場合、感度＞７５％、特異性＞９８％のような情報を含むことができる。当該％表示は、疾患名の蓋然性を示すものではなく、診断の画定した画像に基づく％を表す。感度は該当クラスであると正しく判定される割合であり、特異度は該当のクラスでないと正しく判定される割合ということができる。また、ｆ１－スコアは適合率と再現率の平均を表す。 The diagnostic information can include the f1-score, sensitivity, and specificity of the disease (class). For example, if the disease name is 5, information such as sensitivity >75% and specificity >98% can be included. The percentage display does not indicate the probability of the disease name, but represents the percentage based on the diagnostic defined image. Sensitivity is the rate of correctly determined to be in the corresponding class, and specificity can be said to be the rate of correctly determined not to be in the relevant class. Also, the f1-score represents the average of precision and recall.
　前記疾患診断手段（１３０）の特定の態様としてのFirst Derm（登録商標）により、前記疾患診断手段（１３０）について、更に説明する。 The disease diagnosis means (130) will be further explained by First Derm (registered trademark) as a specific embodiment of the disease diagnosis means (130).
　First Derm（登録商標）については、素人の所有する１０万のスマートフォン皮膚疾患画像を用いたアルゴリズムを開発するために、深層畳み込みニューラルネットワークであるパイトーチ（pytorch）を用いる。この場合、皮膚疾患を、４５及び７０のカテゴリーに分類する標識を用い、前記パイトーチを用いたファスタイ・ライブラリ（Fasti library）を用いて、仮のモデルを作成する。 For FirstDerm (registered trademark), pytorch, a deep convolutional neural network, is used to develop an algorithm using 100,000 smartphone skin disease images owned by amateurs. In this case, markers classifying skin diseases into 45 and 70 categories are used to create a tentative model using the Fasti library using the pie torch.
　前記深層畳み込みニューラルネットワークのモデルは、利用者が前記治療支援装置（１００）に送信した画像に精細調整した３４層のＲｅｓＮｅｔに基づいている。前記モデルは、最適な転移学習を確保するための識別層学習と同様に、迅速な学習時間のための１周期超収束を用いる。前記転移学習（Transfer Learning）は、ある領域で学習したこと（学習済みモデル）を別の領域に役立たせ、効率的に学習させる方法である。 The model of the deep convolutional neural network is based on a 34-layer ResNet fine-tuned to the image that the user sent to the treatment support device (100). The model uses one-cycle super-convergence for fast learning time, as well as discriminative layer learning to ensure optimal transfer learning. Transfer Learning is a method of making learning (learned model) in one area useful in another area for efficient learning.
　前記モデルは、メタデータ及びテキスト入力を取り込むために、同時に入力する方法を用いる。それぞれの最終の標識が十分な数の実例画像を確実に含むようにするために、意味のある粗い標識の多層を有する標識の階層を用いる。クラスの不均衡を処理するために、オーバーサンプリング及び閾値処理を行う。最終的な正確性は、自家製のＴＴＡ技術（test-time augmentation procedure）及びアンサンブル学習により改良する。 The model uses a simultaneous input method to capture metadata and text input. To ensure that each final label contains a sufficient number of example images, we use a label hierarchy with multiple layers of meaningful coarse labels. Oversampling and thresholding are performed to handle class imbalance. The final accuracy is improved by a homegrown TTA technique (test-time augmentation procedure) and ensemble learning.
　前記４５クラスの標識としての疾患は、にきび、日光角化症、血管腫、アトピー性皮膚炎、非定型ほくろ、亀頭炎、基底細胞がん、ライム病、しもやけ、生殖器疣贅、接触性皮膚炎、皮膚のほくろ、皮膚炎、皮膚線維腫、良性の嚢胞、毛嚢炎、癤、性器ヘルペス、単純ヘルペス、帯状疱疹、とびひ、虫刺され、黒子、扁平苔癬、悪性黒色腫、伝染性軟腫症、神経性皮膚炎、黒子（メラノサイト）、真珠状陰茎小丘疹、口囲皮膚炎、バラ色粃糠疹、癜風、炎症後色素沈着過剰症、乾癬、酒皶、フォアダイス斑、脂漏性皮膚炎、脂漏性角化症、扁平上皮癌、梅毒、白癬、蕁麻疹、いぼ、及び白斑である。 Diseases as markers of the 45 classes are acne, actinic keratosis, hemangiomas, atopic dermatitis, atypical moles, balanitis, basal cell carcinoma, Lyme disease, frostbite, genital warts, and contact dermatitis. , skin moles, dermatitis, dermatofibroma, benign cysts, folliculitis, 癤, genital herpes, herpes simplex, herpes zoster, thrush, insect bites, lentigo, lichen planus, malignant melanoma, contagious Molluscum, neurodermatitis, lentigo (melanocytes), pearly penile papules, perioral dermatitis, pityriasis rosea, versicolor, post-inflammatory hyperpigmentation, psoriasis, rosacea, Fordyce macula, seborrheic dermatitis, seborrheic keratosis, squamous cell carcinoma, syphilis, ringworm, urticaria, warts, and vitiligo.
　（３）本発明に係る治療支援システム
　図３に示すように、本発明に係る治療支援システムは、ネットワーク（９００）を介して接続された皮膚疾患の治療支援装置（１００）、利用者端末（２００）、及び任意に店舗管理装置（３００）を含む治療支援システムであって、前記治療支援装置が、前記利用者端末（２００）からの疾患画像及び位置情報を含む利用者情報を受信し、前記利用者端末（２００）に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名及び店舗場所、及び任意に在庫情報を含む、店舗情報を送信する送受信手段（１１０）、前記送受信手段（１１０）から送信される前記疾患画像に基づいて、疾患診断手段（１３０）から前記診断情報を取得し、製品ＤＢ（１２１）から前記診断情報の疾患名に対応する前記製品名を取得し、前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢ（１２２）から取得し、任意に、当該店舗の中から前記製品の在庫を有する店舗及び在庫量を含む前記在庫情報を前記店舗管理装置（３００）から取得し、及び前記疾患画像に応答する前記診断情報、前記製品情報、及び前記店舗情報を前記利用者情報に応じて処理して、前記送受信手段（１１０）に送信する制御手段（１２０）、を備えるシステムであることを特徴とする。 (3) Treatment support system according to the present invention As shown in FIG. 3, the treatment support system according to the present invention includes a skin disease treatment support device (100) connected via a network (900), a user terminal ( 200), and optionally a store management device (300), wherein the treatment support device receives user information including disease images and location information from the user terminal (200), The user terminal (200) includes diagnostic information including disease name, product information including product name responding to the diagnostic information, store name and store location where the product can be obtained, and arbitrarily inventory information. Transmitting/receiving means (110) for transmitting information, based on the disease image transmitted from the transmitting/receiving means (110), the diagnostic information is acquired from the disease diagnosis means (130), and the diagnostic information is obtained from the product DB (121). acquires the product name corresponding to the disease name of the above, and based on the location information of the user information, the store information including the store name and store location of the store near the location and the store from the online store Obtain from the DB (122), optionally from the store management device (300), the inventory information including the store having the product in stock and the amount of inventory in the store, and respond to the disease image. The system is characterized by comprising a control means (120) for processing the diagnostic information, the product information, and the store information according to the user information and transmitting the processed information to the transmitting/receiving means (110).
　（４）本発明の実施形態に係る治療支援方法
　図９に示すように、本発明に係る診断補助方法は、ネットワーク（９００）を介して接続された皮膚疾患の治療支援装置（１００）、利用者端末（２００）、及び任意に、店舗管理装置（３００）を用いる治療支援方法であって、前記治療支援装置（１００）における前記送受信手段（１１０）により、前記利用者端末（２００）からの疾患画像及び位置情報を含む利用者情報を受信する工程（Ｓ１）、前記治療支援装置（１００）における制御手段（１２０）により、前記送受信手段（１１０）から送信される前記疾患画像に基づいて、疾患診断手段（１３０）から前記診断情報を取得する工程（Ｓ２）、前記治療支援装置における制御手段により、製品ＤＢ（１２１）から前記診断情報の疾患名に対応する前記製品名を取得する工程（Ｓ３）、前記治療支援装置における制御手段により、前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢ（１２２）から取得する工程（Ｓ４）、任意に、前記治療支援装置における制御手段により、当該店舗の中から前記製品の在庫を有する店舗及び在庫量を含む前記在庫情報を前記店舗管理装置（３００）から取得する工程（Ｓ５）、及び前記治療支援装置（１００）における前記送受信手段（１１０）により、前記利用者端末（２００）に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名、店舗場所及び在庫情報を含む店舗情報を送信する工程（Ｓ７）、を含む方法であることを特徴とする。 (4) Treatment Support Method According to an Embodiment of the Present Invention As shown in FIG. A treatment support method using a patient terminal (200) and, optionally, a store management device (300), wherein the transmitting/receiving means (110) in the treatment support device (100) receives information from the user terminal (200) A step (S1) of receiving user information including a disease image and position information, based on the disease image transmitted from the transmission/reception means (110) by the control means (120) in the treatment support device (100), A step (S2) of acquiring the diagnostic information from the disease diagnostic means (130), and a step of acquiring the product name corresponding to the disease name of the diagnostic information from the product DB (121) by the control means in the treatment support device ( S3), based on the location information of the user information, store information including store names and store locations of stores near the location and stores from the Internet store is stored in the store DB by the control means in the treatment support device. In the step (S4) obtained from (122), optionally, the control means in the treatment support device transfers the inventory information including stores having inventory of the product from among the stores and the inventory amount to the store management device (300). ), and the transmitting/receiving means (110) in the treatment support device (100), the diagnostic information including the disease name and the product name responding to the diagnostic information are sent to the user terminal (200) and a step (S7) of transmitting store information including the store name, store location, and inventory information where the product can be obtained.
　上記の本発明に係る装置及びシステムの機能構成は、あくまで１例を示すものであり、一つの機能ブロック（データベース及び機能処理部）を分割したり、複数の機能ブロックをまとめて一つの機能ブロックとして構成したりしてもよい。各機能処理部は、コンピュータ装置に内蔵されたＣＰＵ（Central　Processing　Unit）が、ＲＯＭ（Read　Only　Memory）又はハードディスク等の記憶装置に格納されたコンピュータ・プログラムを読み出し、ＣＰＵにより実行されたコンピュータ・プログラムによって実現される。すなわち、各機能処理部は、このコンピュータ・プログラムが、記憶装置に格納されたデータベースやメモリ上の記憶領域からテーブル等の必要なデータを読み書きし、場合によっては、関連するハードウェア（例えば、入出力装置、表示装置、通信インターフェース装置）を制御することによって実現される。また、本発明の実施形態におけるデータベース（ＤＢ）は、商用データベースであってよいが、単なるテーブルやファイルの集合体をも意味し、データベースの内部構造自体は問わないものとする。なお、データベースも１つのサーバ（データベースサーバ）と考えてもよい。 The functional configuration of the apparatus and system according to the present invention described above is merely an example. It may be configured as Each functional processing unit reads a computer program stored in a storage device such as a ROM (Read Only Memory) or a hard disk by a CPU (Central Processing Unit) built into the computer device, and reads the computer program executed by the CPU. realized by That is, each functional processing unit reads and writes necessary data such as a table from a database stored in a storage device and a storage area on a memory, and depending on the case, related hardware (for example, input output device, display device, communication interface device). In addition, the database (DB) in the embodiment of the present invention may be a commercial database, but it also means a mere collection of tables and files, and the internal structure of the database itself does not matter. Note that the database may also be considered as one server (database server).
　以上、実施形態を用いて本発明を説明したが、本発明の技術的範囲は上記実施形態に記載の範囲には限定されないことは言うまでもない。上記実施形態に、多様な変更または改良を加えることが可能であることが当業者に明らかである。またその様な変更または改良を加えた形態も本発明の技術的範囲に含まれ得ることが、請求の範囲の記載から明らかである。 Although the present invention has been described using the embodiments, it goes without saying that the technical scope of the present invention is not limited to the scope described in the above embodiments. It is obvious to those skilled in the art that various modifications or improvements can be made to the above embodiments. Moreover, it is clear from the description of the scope of the claims that the forms with such modifications or improvements can also be included in the technical scope of the present invention.
　１００　　治療支援装置
　１１０　　送受信手段
　１２０　　制御手段
　１２１　　製品ＤＢ
　１２２　　店舗ＤＢ
　１２３　　利用者ＤＢ
　１２４　　医療機関ＤＢ
　１３０　　疾患診断手段
　１３１　　データ群
　１３２　　変換部
　１３３　　ニューラルネットワーク
　２００　　利用者端末
　３００　　店舗管理装置
　４００　　診療予約装置
 
  100 treatment support device 110 transmission/reception means 120 control means 121 product DB
122 store database
123 User DB
124 Medical institution database
130 disease diagnosis means 131 data group 132 conversion unit 133 neural network 200 user terminal 300 store management device 400 medical appointment device

Claims (29)

  1. 　利用者端末とネットワークを介して接続された皮膚疾患の治療支援装置であって、当該治療支援装置が、
    　前記利用者端末からの疾患画像及び位置情報を含む利用者情報を受信し、前記利用者端末に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名及び店舗場所を含む店舗情報を送信する送受信手段、
    　前記送受信手段から送信される前記疾患画像に基づいて、疾患診断手段から前記診断情報を取得し、
    　製品ＤＢから前記診断情報の疾患名に対応する前記製品名を取得し、
    　前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢから取得し、及び
    　前記疾患画像に応答する前記診断情報、前記製品情報、及び前記店舗情報を前記利用者情報に応じて処理して、前記送受信手段に送信する制御手段、
    を備える装置。     A skin disease treatment support device connected to a user terminal via a network, wherein the treatment support device
    User information including disease images and location information is received from the user terminal, and diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and the product are transmitted to the user terminal. Sending/receiving means for transmitting store information including the store name and store location that can be acquired,
    acquiring the diagnostic information from the disease diagnosis means based on the disease image transmitted from the transmission/reception means;
    obtaining the product name corresponding to the disease name of the diagnostic information from the product DB;
    Based on the location information of the user information, store information including store names and store locations of stores near the location and stores from the Internet store is acquired from the store DB, and responding to the disease image. control means for processing the diagnostic information, the product information, and the store information according to the user information and transmitting the information to the transmitting/receiving means;
    A device comprising
  2. 　前記治療支援装置が、さらにネットワークを介して店舗管理装置と接続され、前記送受信手段が、さらに在庫情報を含む店舗情報を送信し、及び
    　前記制御手段が、さらに前記店舗の中から前記製品の在庫を有する店舗及び在庫量を含む在庫情報を店舗管理装置から取得する、［１］に記載の装置。     The treatment support device is further connected to a store management device via a network, the transmitting/receiving means further transmits store information including inventory information, and the control means further controls the inventory of the product from the store. The device according to [1], which acquires inventory information including a store having a store and an inventory amount from a store management device.
  3. 　前記利用者端末の利用者が、患者、一般需要者、医療関係者からなる群から選択されるいずれかの利用者である、請求項１又は２に記載の装置。 　The apparatus according to claim 1 or 2, wherein the user of the user terminal is any user selected from the group consisting of patients, general consumers, and medical personnel.
  4. 　前記利用者端末が、スマートフォン、タブレット端末、パソコンからなる群から選択されるいずれかである、請求項１～３のいずれか１項に記載の装置。 The device according to any one of claims 1 to 3, wherein the user terminal is one selected from the group consisting of smartphones, tablet terminals, and personal computers.
  5. 　前記疾患画像が、患部に焦点をあてた、スマートフォンカメラ撮影による画像又は画像ライブラリからの画像である、請求項１～４のいずれか１項に記載の装置。 　The apparatus according to any one of claims 1 to 4, wherein the disease image is an image taken with a smartphone camera or an image from an image library, focusing on the affected area.
  6. 　前記位置情報の位置が、前記利用者の現在位置又は前記利用者の指定位置である、請求項１～５のいずれか１項に記載の装置。 The device according to any one of claims 1 to 5, wherein the position of the position information is the user's current position or the user's designated position.
  7. 　前記疾患名が１又は複数の疾患候補名で提供される、請求項１～６のいずれか１項に記載の装置。 The device according to any one of claims 1 to 6, wherein the disease name is provided as one or more disease candidate names.
  8. 　前記製品が、医療用医薬品、一般用医薬品、並びにトクホ及び機能性表示食品を含む機能性食品からなる群から選択される、請求項１～７のいずれか１項に記載の装置。 The device according to any one of claims 1 to 7, wherein the product is selected from the group consisting of ethical drugs, over-the-counter drugs, and functional foods including FOSHU and functionally labeled foods.
  9. 　前記店舗が、薬局又はドラッグストアである、請求項１～８のいずれか１項に記載の装置。 The device according to any one of claims 1 to 8, wherein the store is a pharmacy or drug store.
  10. 　前記製品ＤＢが、市販又は自家製の、医療用医薬品、一般用医薬品、並びにトクホ及び機能性表示食品を含む機能性食品からなる群から選択される少なくとも1つを含むＤＢである、請求項１～９のいずれか１項に記載の装置。 Claims 1-, wherein the product DB is a DB containing at least one selected from the group consisting of commercially available or homemade medical drugs, over-the-counter drugs, and functional foods including FOSHU and functionally labeled foods. 10. Apparatus according to any one of 9.
  11. 　前記疾患診断手段が、前記治療支援装置の一部分である又は前記治療診断装置とネットワークを介して接続されている装置である、請求項１～１０のいずれか１項に記載の装置。 The device according to any one of claims 1 to 10, wherein said disease diagnosis means is a part of said treatment support device or a device connected to said treatment diagnosis device via a network.
  12. 　前記在庫管理装置が、前記治療支援装置の一部分である又は前記治療診断装置とネットワークを介して接続されている、請求項２～１１のいずれか１項に記載の装置。 The device according to any one of claims 2 to 11, wherein said inventory management device is part of said treatment support device or is connected to said treatment diagnosis device via a network.
  13. 　前記在庫管理装置が、一店舗の在庫管理装置又は複数の店舗の在庫管理装置である、請求項２～１２のいずれか１項に記載の装置。 The device according to any one of claims 2 to 12, wherein the inventory management device is an inventory management device for one store or an inventory management device for a plurality of stores.
  14. 　前記制御手段が、さらに利用者ＤＢを備え、
    　前記制御手段が前記送受信手段を通して、前記利用者端末からの利用者登録を受けつけ、利用者登録情報を前記制御手段の利用者ＤＢに記録し、前記利用者端末に利用者ＩＤ及びパスワードを返信する、請求項１～１３のいずれか１項に記載の装置。     The control means further comprises a user DB,
    The control means receives user registration from the user terminal through the transmission/reception means, records user registration information in a user DB of the control means, and returns a user ID and password to the user terminal. A device according to any one of claims 1 to 13.
  15. 　前記制御手段が前記送受信手段を通して、前記利用者端末からの製品発注及び／又は発送の依頼を受けつけ、前記店舗管理装置に製品発注及び／又は発送の依頼情報を送信する、請求項１～１４のいずれか１項に記載の装置。 Claims 1 to 14, wherein said control means receives a request for product ordering and/or shipping from said user terminal through said transmitting/receiving means, and transmits request information for product ordering and/or shipping to said store management device. A device according to any one of the preceding claims.
  16. 　前記制御手段が、さらに医療機関ＤＢを備え、医療機関の診療予約装置とネットワークを介して接続され、
    　前記製品情報における製品が医療用医薬品であるとき、前記利用者情報に基づいて皮膚疾患の専門医師が診療する医療機関を医療機関ＤＢに問合せして、医療機関情報を取得し、前記利用者端末に送信する、請求項１～１５のいずれか１項に記載の装置。     the control means further comprises a medical institution DB and is connected to a medical appointment device of the medical institution via a network;
    When the product in the product information is an ethical drug, a medical institution DB is inquired about a medical institution where a skin disease specialist treats based on the user information, and the medical institution information is acquired, and the user terminal 16. A device according to any one of claims 1 to 15, adapted to transmit to.
  17. 　前記制御手段が、さらに前記診療予約装置から診療予約情報を取得して、前記利用者端末に送信し、
    前記利用者端末から送信される診療希望情報を前記診療予約装置に送信して、前記診療予約装置から送信される受付情報を前記利用者端末に送信する、請求項１６に記載の装置。     The control means further acquires medical appointment information from the medical appointment device and transmits it to the user terminal,
    17. The apparatus according to claim 16, which transmits medical treatment request information transmitted from said user terminal to said medical treatment reservation device, and transmits reception information transmitted from said medical treatment reservation device to said user terminal.
  18. 　前記疾患診断手段が、疾患名の標識を付された疾患画像のデータ群、各疾患画像を特徴量データに変換する変換部、前記データ群で機械学習処理が施されたニューラルネットワークモデルを備え、
    　前記制御手段から送信された診断対象の疾患画像を前記変換部で特徴量データに変換し、前記ニューラルネットワークモデルにより計算して特徴量データに変換し、前記標識を識別して疾患名を含む診断情報を取得し、前記制御手段に送信する、
    請求項１～１７のいずれか１項に記載の装置。     The disease diagnosis means comprises a data group of disease images labeled with disease names, a conversion unit that converts each disease image into feature amount data, and a neural network model subjected to machine learning processing on the data group,
    The disease image to be diagnosed transmitted from the control means is converted into feature amount data by the conversion unit, calculated by the neural network model, converted into feature amount data, and the signs are identified to make a diagnosis including the name of the disease. obtaining information and sending it to the control means;
    A device according to any one of claims 1-17.
  19. 　前記ニューラルネットワークモデルが、深層畳み込みニューラルネットワークモデルである、請求項１８に記載の装置。 The apparatus according to claim 18, wherein said neural network model is a deep convolutional neural network model.
  20. 　前記データ群が、５０，０００から４００，０００の疾患画像を含む、請求項１８又は１９に記載の装置。 20. The apparatus according to claim 18 or 19, wherein said data set comprises 50,000 to 400,000 disease images.
  21. 　前記深層畳み込みニューラルネットワークが、３４層のＲｅｓＮｅｔ（Residual Network）モデルである、請求項１９又は２０に記載の装置。 The apparatus according to claim 19 or 20, wherein said deep convolutional neural network is a 34-layer ResNet (Residual Network) model.
  22. 　前記疾患名が、ＩＣＤコードの疾患カテゴリーで表される、［１８］～［２１］のいずれか１項に記載の装置。 The device according to any one of [18] to [21], wherein the disease name is represented by the disease category of the ICD code.
  23. 　前記疾患名が、４５又は７０の疾患カテゴリーで表される、請求項１８～２２のいずれか１項に記載の装置。 The device according to any one of claims 18 to 22, wherein the disease name is represented by 45 or 70 disease categories.
  24. 　前記診断情報の疾患名が、１～５種のいずれか１つの疾患カテゴリーの疾患候補名として提示される、請求項１８～２３のいずれか１項に記載の装置。 The apparatus according to any one of claims 18 to 23, wherein the disease name of the diagnostic information is presented as a disease candidate name of any one of 1 to 5 disease categories.
  25. 　前記診断情報が、ｆ１－スコア、感度、及び特異性の数値を含む、請求項１８～２４のいずれか１項に記載の装置。 The device according to any one of claims 18 to 24, wherein said diagnostic information includes numerical values of f1-score, sensitivity and specificity.
  26. 　ネットワークを介して接続された皮膚疾患の治療支援装置及び利用者端末を備える治療支援システムであって、
    　前記治療支援装置が、
    　前記利用者端末からの疾患画像及び位置情報を含む利用者情報を受信し、前記利用者端末に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名及び店舗場所を含む店舗情報を送信する送受信手段、
    　前記送受信手段から送信される前記疾患画像に基づいて、疾患診断手段から前記診断情報を取得し、
    　製品ＤＢから前記診断情報の疾患名に対応する前記製品名を取得し、
    　前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢから取得し、及び
    　前記疾患画像に応答する前記診断情報、前記製品情報、及び前記店舗情報を前記利用者情報に応じて処理して、前記送受信手段に送信する制御手段、
    を備える、システム。     A treatment support system comprising a skin disease treatment support device and a user terminal connected via a network,
    The treatment support device is
    User information including disease images and location information is received from the user terminal, and diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and the product are transmitted to the user terminal. Sending/receiving means for transmitting store information including the store name and store location that can be acquired,
    acquiring the diagnostic information from the disease diagnosis means based on the disease image transmitted from the transmission/reception means;
    obtaining the product name corresponding to the disease name of the diagnostic information from the product DB;
    Based on the location information of the user information, store information including store names and store locations of stores near the location and stores from the Internet store is acquired from the store DB, and responding to the disease image. control means for processing the diagnostic information, the product information, and the store information according to the user information and transmitting the information to the transmitting/receiving means;
    A system comprising:
  27. 　前記治療支援装置が、さらにネットワークを介して店舗管理装置と接続され、前記送信手段が、さらに在庫情報を含む店舗情報を送信し、及び
    　前記制御手段が、さらに前記店舗の中から前記製品の在庫を有する店舗及び在庫量を含む在庫情報を店舗管理装置から取得する、請求項２６に記載のシステム。     The treatment support device is further connected to a store management device via a network, the transmission means further transmits store information including inventory information, and the control means further controls the inventory of the product from the store. 27. The system according to claim 26, wherein the inventory information including the store having the and the inventory amount is obtained from the store management device.
  28. 　ネットワークを介して接続された皮膚疾患の治療支援装置及び利用者端末を備えるシステムを用いる治療支援方法であって、
    　前記治療支援装置における送受信手段により、前記利用者端末からの疾患画像及び位置情報を含む利用者情報を受信する工程、
    　前記治療支援装置における制御手段により、前記送受信手段から送信される前記疾患画像に基づいて、疾患診断手段から診断情報を取得する工程、
    　前記治療支援装置における制御手段により、製品ＤＢから前記診断情報の疾患名に対応する製品名を取得する工程、
    　前記治療支援装置における制御手段により、前記利用者情報の前記位置情報に基づいて、当該位置の近傍の店舗及びネット店舗からの店舗の、店舗名及び店舗場所を含む店舗情報を店舗ＤＢから取得する工程、
    　前記治療支援装置における前記送受信手段により、前記利用者端末に、疾患名を含む診断情報、当該診断情報に応答した製品名を含む製品情報、並びに前記製品を取得できる店舗名及び店舗場所を含む店舗情報を送信する工程、
    を含む、方法。     A treatment support method using a system comprising a skin disease treatment support device and a user terminal connected via a network,
    a step of receiving user information including a disease image and location information from the user terminal by means of transmission/reception means in the treatment support apparatus;
    a step of acquiring diagnostic information from a disease diagnosis means based on the disease image transmitted from the transmission/reception means by the control means in the treatment support apparatus;
    obtaining a product name corresponding to the disease name in the diagnostic information from the product DB by means of control means in the treatment support device;
    Based on the location information of the user information, store information including the name and location of the store near the location and the store from the online store is acquired from the store DB by the control means of the treatment support device. process,
    Diagnostic information including a disease name, product information including a product name responding to the diagnostic information, and a store including a store name and a store location where the product can be obtained are sent to the user terminal by the transmitting/receiving means of the treatment support device. transmitting the information;
    A method, including
  29. 　前記治療支援装置が、ネットワークを介して、さらに店舗管理装置と接続され、
    　前記治療支援装置における制御手段により、さらに前記店舗の中から前記製品の在庫を有する店舗及び在庫量を含む在庫情報を前記店舗管理装置から取得する工程、及び
    　前記治療支援装置における前記送受信手段により、前記利用者端末に、さらに前記在庫情報を含む店舗情報を送信する工程、
    を含む、請求項２８に記載の方法。
     
          The treatment support device is further connected to a store management device via a network,
    a step of acquiring inventory information including a store having inventory of the product and an inventory amount from the store management device by the control means of the treatment support device; a step of further transmitting store information including the stock information to the user terminal;
    29. The method of claim 28, comprising:
PCT/JP2021/036999 2021-10-06 2021-10-06 Treatment assistance device, system, and method WO2023058157A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/036999 WO2023058157A1 (en) 2021-10-06 2021-10-06 Treatment assistance device, system, and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/036999 WO2023058157A1 (en) 2021-10-06 2021-10-06 Treatment assistance device, system, and method

Publications (1)

Publication Number Publication Date
WO2023058157A1 true WO2023058157A1 (en) 2023-04-13

Family

ID=85803319

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/036999 WO2023058157A1 (en) 2021-10-06 2021-10-06 Treatment assistance device, system, and method

Country Status (1)

Country Link
WO (1) WO2023058157A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002007568A (en) * 2000-06-27 2002-01-11 Kaihatsu Komonshitsu:Kk Diagnostic system, diagnostic data generating method, information processing apparatus used for them, terminal device, and recording medium
JP2010515489A (en) * 2007-01-05 2010-05-13 マイスキン インコーポレイテッド System, apparatus and method for imaging skin
JP2015530886A (en) * 2012-06-27 2015-10-29 バウチャー、ライアン Apparatus, method and system for obtaining medical diagnostic information, and provision of telemedicine services
JP2018038789A (en) * 2016-09-02 2018-03-15 カシオ計算機株式会社 Diagnosis support device, method for processing image on diagnosis support device and program

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002007568A (en) * 2000-06-27 2002-01-11 Kaihatsu Komonshitsu:Kk Diagnostic system, diagnostic data generating method, information processing apparatus used for them, terminal device, and recording medium
JP2010515489A (en) * 2007-01-05 2010-05-13 マイスキン インコーポレイテッド System, apparatus and method for imaging skin
JP2015530886A (en) * 2012-06-27 2015-10-29 バウチャー、ライアン Apparatus, method and system for obtaining medical diagnostic information, and provision of telemedicine services
JP2018038789A (en) * 2016-09-02 2018-03-15 カシオ計算機株式会社 Diagnosis support device, method for processing image on diagnosis support device and program

Similar Documents

Publication Publication Date Title
US11735294B2 (en) Client management tool system and method
US10937164B2 (en) Medical evaluation machine learning workflows and processes
Dick et al. Accuracy of computer-aided diagnosis of melanoma: a meta-analysis
Pacis et al. Trends in telemedicine utilizing artificial intelligence
US20170011192A1 (en) Patient And Mobile Healthcare Worker Initiated Encounter In A Telemedicine System
US20160103963A1 (en) Method and system for smart healthcare management
EP3255573A1 (en) Clinical decision supporting ensemble system and clinical decison supporting method using the same
Quer et al. Augmenting diagnostic vision with AI
CN102542127A (en) Systems and methods for smart medical collaboration
US11328815B2 (en) Physical measurement of empirical indicators of patient-related outcome value using time and motion sensor results
EP3910648A1 (en) Client management tool system and method
US20230133330A1 (en) Systems and methods for medical resource intelligence
US20170357770A1 (en) Health risk cloud analysis system capable of integrating gene information and acquired lifestyle
Mersini et al. APPification of hospital healthcare and data management using QRcodes
Vyas et al. Smart health systems: emerging trends
KR20140037325A (en) Telemedicine system and method based ecg data using bio- signal meter and smart device
US11200967B1 (en) Medical patient synergistic treatment application
Choudhuri et al. Internet of Things in healthcare: a brief overview
WO2023058157A1 (en) Treatment assistance device, system, and method
CN112749321A (en) Data processing method, client, server, system and storage medium
CN112750512A (en) Data processing method, client, server, system and storage medium
Sudhakara et al. A technology framework for remote patient care in dermatology for early diagnosis
Krishnasamy et al. An innovative outcome of internet of things and artificial intelligence in remote centered healthcare application schemes
CN112842302A (en) Method and system for processing body information and storage medium
JEMAA et al. Digital Twin For A Human Heart Using Deep Learning and Stream Processing Platforms

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21959899

Country of ref document: EP

Kind code of ref document: A1